Abstract: Over the past 40 years Beck’s cognitive theory of depression has emerged as one of the major cognitive theories to explain depressive phenomena (Oei & Dingle, 2008). It has been proposed that two levels of cognitions, that is, dysfunctional attitudes and automatic thoughts, contribute to the development and maintenance of depression. However, little empirical research has been carried out to specify and examine the differential roles of dysfunctional attitudes and automatic thoughts in the development of depression. In a comprehensive review Oei and Kwon (2007) creatively proposed the Integrated Cognitive model of depression (ICM) to test the differential roles of dysfunctional attitudes and automatic thoughts. Whereas ICM has been studied in western adult populations, the research examining this model in adolescents or non-western population lagged far behind. Longitudinal cross-cultural evidence are needed to evaluate the validity of the ICM in Chinese adolescents. With the two-wave panel data we hypothesized that in the relationship between adolescents’ negative life events experienced between Time1 and Time2 (ASLEC2) and changes in the severity of depressive symptoms (CES-D2–CES-D1) dysfunctional attitudes measured at Time1(DAS1) served as a cognitive moderator and the frequency of automatic thoughts between Time1 and Time2 (ATQ2–ATQ1) as a cognitive mediator. That was dysfunctional attitudes was a mediated moderator. Additionally the model did not exclude the possibility that either negative life events or dysfunctional attitudes might have some direct influences on automatic thoughts and depressive symptoms and was called Integrated Cognitive Model of Depression for Adolescents (ICM-A). As a contrast we also tested the integrated cognitive model (ICM) and the symptom model (SM). Participants were 613 Chinese junior high school students. All participants completed the Adolescent Life Events Questionnaire (ASLEC), Dysfunctional Attitudes Scale (DAS), Automatic Thoughts Questionnaire (ATQ) and Center for Epidemiological Studies Depression Scale (CES-D) twice -about 4 months apart. The results of the path analysis indicated : (1) The correlations among the measures were significant and repeated measures t-tests revealed that all the measures increased significantly from Time1 to Time2; (2) The modified ICM-A best model fitted the Chinese adolescent data and there was a gender difference in the path estimate from the interaction between dysfunctional attitudes and negative life events to automatic thoughts. We concluded that at the increasing phase of adolescents’ depressive symptoms dysfunctional attitudes could be a common cognitive moderator of depression, whereas automatic thoughts could be specific cognitive mediators of depression and there was a gender difference in the model. The findings support the moderating role of dysfunctional attitudes and the mediating role of automatic thoughts in the relationships between negative life events and depressive symptoms at the increasing phase of adolescents’ depressive symptoms. This indicates the main cognitive constructs of Chinese adolescents, such as the roles of dysfunctional attitudes and automatic thoughts, may share similarities when compared to Western populations. This provides initial support that Western-based theories and treatments for depression may be suitable for use in some non-Western populations. Cognitive intervention will be effective in Chinese adolescents’ depression prevention. And the current investigation provides preliminary evidence of the gender difference in the moderating role of dysfunctional attitudes.

Key words: adolescents, depression, dysfunctional attitudes, automatic thoughts, gender difference